Ambient Temperature Effects on User Thermal ...

Proceedings of the Human Factors and Ergonomics Society 59th Annual Meeting - 2015

1090

Ambient temperature effects on user thermal sensation with a simulated tablet computer Han Zhang1, Alan Hedge1, Beiyuan Guo2 Department of Design and Environmental Analysis, Cornell University College of Human Ecology, MVR Hall, Ithaca, NY 14853 2 State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University Beijing, China 1

A tablet computer’s surface temperature can reach levels that can lead to user discomfort, especially in a warm environment. The ambient environments in which tablet computers are used can also vary. To understand how users perceive the heat from tablet computers, a laboratory study was conducted with controlled surface temperatures and ambient temperatures. A positive relationship between surface temperature and participants’ thermal sensation scores was found. Participants’ thermal responses to the surface heat of a simulated tablet were also moderated by the indoor temperature. Higher surface temperature (44°C) was rated less warm in cool environment than hot environment, while lower surface temperatures (34-38°C) were rated warmer in cool than hot environment. The thermal responses corresponding to the tablet surface temperatures and ambient temperatures will be helpful for setting future tablet computer heat dissipation design limits.

Copyright 2015 Human Factors and Ergonomics Society. DOI 10.1177/1541931215591156

INTRODUCTION Tablet computer surface temperatures have been reported to rise from room temperature up to 47°C in a warm room (Chatterjee, 2014; Tapellini, 2012). Skin contact with a warm or hot hand-held device surface might cause users’ thermal discomfort and increase skin burn risks. Fingertips and the palm areas typically are in contact with a tablet computer surface (Zhang, Hedge & Guo, 2015). Previous laboratory studies have determined the thermal sensation thresholds for the fingers and palm areas. The warm sensation threshold for the index finger is 34.8 (±2.21)°C in a room temperature of 25°C (Hirosawa et al., 1984), and for middle finger is 35.6 (±2.6)°C (Toibana et al., 1993), and the threshold for the thenar eminence is between 32.5 to 34.7°C, varying among studies (Harju, 2002; Kelly et al., 2005). In the above studies the tested duration of skin contact ranged from 3 seconds (Hirosawa et al., 1984) to 12 seconds (Kelly et al., 2005). Since the use of a tablet often requires holding the device for prolonged periods of time with multiple fingers and palm skin areas in contact with the bottom surface of the tablet, a better understanding of users’ thermal sensation for these areas is needed. Both the stimulus duration and stimulus area size can affect users’ thermal sensation. The heat pain threshold tends to decrease with larger areas of heat stimulation and with increased duration (Dyck et al., 1993; Meh & Denislic, 1994; Pertovaara et al., 1996; Hagander et al., 2000). Ambient temperature can also affect the human perception of thermal stimuli (Hirosawa et al., 1984; Strigo et al., 2000; Harvey et al., 2012). People perceive less intensity for cold (0°C -25°C) and hot (44°C -50 °C) stimuli in cool ambient temperature of 15°C but more intensity and unpleasantness for hot stimuli in warm environment of 35°C, although no such effects were found for the warm stimuli of 37°C and 40°C (Strigo et al., 2000).

However, an environmental temperature effect is not consistent in all the studies. Lower ambient temperatures of 15°C have been reported to have suppressive effect on heat pain threshold (Strigo et al., 2000). The response time to noxious heat stimuli of 60°C was increased at an ambient temperature of 10°C in a rat study, indicating that the heat pain threshold was increased with the low ambient temperature (Schoenfeld et al., 1985). On the contrary, Croze et al. (1977) and Pertovaara et al. (1996) reported skin temperature did not affect the heat pain threshold. Ambient temperature has been shown to increase warm sensation thresholds at the fingers by increasing skin temperature (Hirosawa et al., 1984), and similarly skin temperature affects innocuous thermal sensation thresholds in other human studies (Kojo and Pertovaara, 1987; Molinari et al., 1977) and animal studies (Greenspan and Kenshalo, 1985; Sumino and Dubner, 1981). However, no significant effects of environmental temperatures were found on reported thermal sensation for warm stimuli of 37°C and 40°C (Strigo et al., 2000). The effect of the ambient environment on thermal stimuli was also investigated in recent computer related studies (Baugh and Doherty, 2011; Harvey et al., 2012). Harvey et al. (2012) tested the thenar eminence and the back of wrist for thermal feedback as mobile device notification. Environmental temperature was not controlled but recorded and it ranged from 8.45°C to 27.75°C. In general, ambient temperature had a significant effect on the user’s thermal stimuli detection time and perceived comfort. In the range of 15-20°C users experienced the greatest comfort with thermal stimuli of 16°C to 38°C. However, in a laptop surface temperature study, no significant effect on thermal comfort was found between an environment at 23°C and 35°C (Baugh and Doherty, 2011). Tablet computers can be used in a variety of ambient environments, and ambient temperature has been shown to affect thermal sensation though previous research on its effect on non-noxious warm stimuli is inconsistent. In addition, in previous cutaneous thermal sensation laboratory studies the

Downloaded from pro.sagepub.com at CORNELL UNIV on May 18, 2016


skin-stimulus contact area was commonly around 10mm2 with a probe touching the skin for a short duration under 15 seconds. Thus, these results cannot accurately predict tablet computer users’ real thermal responses. Therefore, it is necessary to understand how users’ thermal sensation and comfort change towards a heated tablet surface with different indoor temperatures. METHODS Participants In total seventy-five participants were recruited from students and employees in Cornell University. Thirty-three participants were male and forty-two were female. The participants’ ages range from 18 to 64, with an average age of 27.7 years and a standard deviation of 8.6 years. Participants with previous upper extremity injuries, neurological disorders, or diabetes were excluded from the study, to preclude possible variations in thermal sensation.

1091

they would normally hold a tablet computer for 90 seconds, and to report their thermal sensation and thermal comfort 3 times at each temperature level (0, 45 and 90 seconds). Two visual analog scales of thermal sensation and thermal comfort were used and so that they did not have to release their grip participants verbally made their ratings to the experimenter. The scales range from 0-100, with 50 as neutral. The thermal sensation scale ranged from extremely cold (score 0) to extremely hot (score 100). The thermal comfort scale ranged from extremely comfortable (score 0) to extremely uncomfortable (score 100). Between each temperature level, the participant put both their hands on a controlled temperature surface of 32°C for 30 seconds and rested them in the air for 2 minutes. In total, each test session lasts 55-60 minutes. The experiments were conducted from December 2014 to January 2015. The study protocol was approved by the Cornell University Institutional Review Board.

Apparatus All experimental sessions were conducted in a controlled environmental chamber. Indoor air temperature was controlled at three levels of 13°C, 23°C and 33°C, while the humidity was controlled at 40% RH. A simulated tablet computer, 24.4 x 18.5 cm (9.6 x 7.3 inches), with a heating surface was developed as shown in Figures 1 and 2. The prototype comprised nine 5.1 x 2.5 cm rectangular heating pads connected with heaters (Kapton 28 Volts, 20 Watt) and thermal sensors controlled by National Instrument LabView (version 13.0.1f2, 32-bit) proportional-integral-derivative (PID) module. The frame was ABS plastic and the heating pads were aluminum. This system allows the control of surface temperatures at different levels, ranging from 34°C to 44°C. An infrared camera (FLIR E30) was used to measure participant’s initial hand surface temperatures.

Figure 1. A participant holding the heating surface

Procedure Participants spent about 20 minutes in the laboratory acclimating to the environment before the start of experiment. During this time information was gathered on their age, weight and height. Participants’ hand dimensions were measured with a ruler. The researcher photographed the participants’ hands holding the surface from the bottom of the prototype, and measured the initial palm skin surface temperature with the IR camera. Two independent variables were tested, including surface temperature and indoor temperature. The surface temperature was controlled from 34°C to 44°C, with 2°C as an interval. In total 6 levels were tested, and each level was tested twice. The order of tested temperatures were counterbalanced. The tested temperatures simulate the range of tablet computer temperatures from close to skin temperature to under 45°C, which is the skin burn threshold for prolonged contact of 8 hours and longer (ISO 13732-1:2006). For each temperature level, participants were asked to hold the prototype in the way

Figure 2. Infrared (IR) image of the heating surface Data analysis The data were analyzed using statistical software (JMP 10.0.0). Two-way repeated analysis of variance (ANOVA) was used to test the significance of the main effects and interactions of surface temperature and indoor temperatures for thermal sensation and comfort. The variables of holding duration, gender, height, weight, and hand size measurements were also tested. Tukey’s HSD procedure was used for posthoc analysis to test significance between different levels

Downloaded from pro.sagepub.com at CORNELL UNIV on May 18, 2016


1092

within a main effect or interaction. P